The growing role of methane in anthropogenic climate change

Atmospheric methane (CH₄) has experienced puzzling dynamics over the past 15 years. After a period of relative stagnation in the early 2000 s (+0.5 ±3.1 ppb yr⁻¹ increase on average for 2000–2006), atmospheric methane concentrations have increased rapidly since 2007 at more than ten times this rate (+6.9 ± 2.7 ppb yr⁻¹ for 2007–2015; figure 1 top left; Dlugokencky 2016). The atmospheric growth rate of methane accelerated to +12.5 ppb in 2014 and +9.9 ppb in 2015, reaching an annual average concentration of 1834 ppb in 2015 (Dlugokencky 2016). Because of this acceleration, the evolution of atmospheric methane over the last three years is inconsistent with the mitigation required in the Representative Concentration Pathways (RCP) of 2.5, 4 and 6 W m⁻² and now most closely aligns with the RCP 8.5 W m⁻² (figure 1 top left) (Fujino et al 2006, Clarke et al 2007, Riahi et al 2007, van Vuuren et al 2007). This emerging dynamic highlights methane's growing contribution to global warming relative to the observed slower growth rates of CO₂ over the past three years (Le Quéré et al 2016, ESSD; figure 1 top right, Jackson et al 2016) and a relatively constant growth rate of nitrous oxide (N₂O) (Hartmann et al 2013).

Zoom In Zoom Out Reset image size

The balance of surface sources and sinks determines the global methane budget. Surface sources include methane originating from biogenic (wetlands, lakes, agriculture, waste/landfill, permafrost), thermogenic (fossil fuel usage and natural seeps), pyrogenic (biomass and biofuel burning) or mixed (hydrates, geological) sources. Dominant sinks include methane oxidation by the hydroxyl radical (OH) and other radicals in the atmosphere as well as methanotrophy in soils. Based on a new ensemble of atmospheric studies, global emissions are estimated at 559 [540–568] Tg CH₄.yr⁻¹ for the 2003–2012 decade (Saunois et al 2016). Tropical sources, including both natural and anthropogenic sources represent two-thirds of total global emissions and are dominated by emissions from wetlands (figure 2). Approximately two-thirds of global emissions are also attributable to anthropogenic activities, including those from both mid-latitudes and the tropics (e.g., agriculture and waste, figure 2).

Zoom In Zoom Out Reset image size

Despite substantial knowledge about the location, size and trends of methane sources and sinks, the relative contributions explaining the recent atmospheric increase remain uncertain (e.g. Nisbet et al 2014, 2016). Based on activity data and emission factors from various anthropogenic sectors, bottom-up inventories of anthropogenic emissions estimate an increase of fossil-related emissions of 3–4 Tg each year since 2007 (EPA 2012; EDGAR 2014). Using ethane measurements and methane-to-ethane ratios, Haussmann et al (2016) also suggest a substantial contribution of fossil-related emissions (18%–73% of the total increase in atmospheric methane). ¹³CH₄ isotopic observations show a significant depletion of ¹³C in the atmosphere (∼−0.12‰ in seven years), suggesting that increases in methane emissions after 2006 are primarily biogenic and are more consistent with sources from agriculture than natural wetlands (Nisbet et al 2016; Schaefer et al 2016). Recent bottom-up inventories estimate an increase in agricultural annual emissions of 3–5 Tg between 2006 and 2012, mostly from Africa and Asia, whereas wetland emissions were estimated to be mostly unchanged between 2006 and 2012 (Poulter et al 2016). Meanwhile, biomass burning emissions decreased by 2–3 Tg yr⁻¹ between 2007 and 2013 compared to 2000 and 2006, although the recent El Niño conditions have lead to abnormally large peat fires in Indonesia (Van der Werf et al 2016). Not accounting for this long-term decrease in the ¹³C-heavy methane source from biomass burning, and based on ¹³C atmospheric observations and on an enriched database for isotopic source signatures, Schwietzke et al (2016) even find decreasing fossil fuel emissions since 2000, a different conclusion than reached in most other recent studies.

Sinks may also be playing a role in the rapid rise in atmospheric methane over the last decade (figure 1). Using a chemistry-transport model run over 40 years, Dalsøren et al (2016) infer a stabilization of OH concentrations after 2006, in contrast to a total 3% increase since the late 1990s (8% since the 1970s). Stabilized OH concentrations can increase methane lifetimes and may help explain the atmospheric methane increase as well, as a decrease of 1% in atmospheric OH concentrations is roughly equivalent to ∼5 Tg yr⁻¹ of increased methane emissions (e.g. Saunois et al 2016).

These various factors notwithstanding, there is no consensus scenario of methane sources and sinks that explains the atmospheric increase since 2007 (Kirschke et al 2013). Recent evidence from atmospheric observations suggests three main contributors for emission changes. The first element is an increase in biogenic emissions, mostly from agriculture (¹³C compatible, Schaefer et al 2016). The second is an increase of fossil-related emissions (ethane-compatible, Haussman et al 2016). The third is a decrease of biomass burning emissions (¹³C compatible and cancelling a fossil fuel increase, Van der Werf et al 2016). The necessity of an anthropogenic emission increase can still be reduced by a possible stagnation of OH concentrations or by regional contributions from wetland emissions, such as emissions fluctuations resulting from drought conditions in South America (e.g., in 2010–2011; Basso et al 2016).

At the regional scale, methane emissions contributing to the observed atmospheric increase since 2006 are most likely tropical, although some mid-latitude regions, such as China, also appear to contribute to the increase (e.g. Bergamaschi et al 2013). To date, no significant contribution to the atmospheric increase from Arctic regions has been found, except in 2007 and attributable then to a relatively warm and late summer (Dlugokencky et al 2011). Contrary to a recent estimate based on three different atmospheric inversions (Turner et al 2016), no trend in US methane emissions is found in the ensemble of inversions gathered in Saunois et al (2016), and thus a substantial contribution of US shale gas industry to the recent methane atmospheric increase seems unlikely (Bruhwiler et al 2016).

Scientific breakthroughs are needed to predict methane emissions today and in the future, particularly with a changing climate. First, annual to decadal CH₄ emissions from natural wetlands and other inland water systems are highly uncertain. The sum of all natural methane sources as inferred by process-based bottom-up modelling is too large by about 30% compared to the constraint provided by methane atmospheric mixing ratios. The strategy to address this issue requires developing and synthesizing (i) direct methane flux measurements in the field to constrain the parametrizations of land surface models similarly to Fluxnet-CO₂, (ii) process-based models for lakes, rivers, and permafrost methane emissions (e.g., Tan and Zhuang 2015 for lakes), and (iii) dynamic global high resolution maps (50–100 m) with all inland water surfaces consistently categorized to avoid double counting emitting surfaces (Yamazaki et al 2015).

Second, the partitioning of CH₄ emissions and sinks by region and process needs to be better constrained by atmospheric observations and process-based models. Beyond the recurring need for a broader network of methane observations, it is essential (i) to extend observations of tracers more specific to individual methane sources and sinks such as methane isotope concentrations and emission signatures (Röckmann et al 2011; Schaefer et al 2016) and ethane (Haussmann et al 2016), and (ii) to improve the estimation of magnitude and trend of OH radicals by better quantifying its sources and sinks in chemistry climate models (e.g. Dalsøren et al 2016). The latter will benefit from a recent inter-comparison of chemistry climate models (the Chemistry Climate Model Initiative (CCMI) update of Lamarque et al 2013) and CMIP6 simulations scheduled for the next IPCC report. Breakthrough technologies already allow high precision measurements of methane and its isotopes at the surface, for instance using cavity ring down spectrometers such as in Maher et al (2014). Future LIDAR measurements from space will provide the first low-bias global estimate of methane atmospheric columns all year round beginning in ∼2020 (Kiemle et al 2014). The partitioning of emissions will also benefit from efforts to improve and regularly update anthropogenic inventories.

Third, uncertainties in the modelling of atmospheric transport and chemistry limit the optimal assimilation of atmospheric observations and increase the uncertainties of the inversion-derived flux estimates. Key steps should include the improvement of OH fields and other methane sinks (e.g., methane oxidation by other radicals), refinements in the horizontal and vertical model grids, parameterization of vertical mixing and representation of stratospheric concentrations. Such modelling improvements could be accomplished through regular inter-comparisons such as TRANSCOM (e.g., Patra et al 2011) or CCMI (Lamarque et al 2013) and through additional efforts for model validation (Bergamaschi et al 2013).

Despite important uncertainties in methane sources and sinks, the recent increase in methane concentrations suggests a dominant anthropogenic contribution (either biogenic or thermogenic). Methane therefore offers growing opportunities for climate change mitigation that could allow a return to lower emission trajectories such as RCP6 or RCP4.5. Because of methane's high global warming potential and short lifetime in the atmosphere compared to CO₂, its mitigation offers the possibility to slow climate change efficiently in a shorter time horizon. In addition to climate benefits, reducing methane emissions could help improve human health and crop production through simultaneous reductions in ozone production (West et al 2013; Shindell 2016) and provide business and employment opportunities. A diverse set of strategies already exists, as proposed by multilateral partnerships such as the Global Methane Initiative (www.globalmethane.org) and the Climate and the Clean Air Coalition (www.ccacoalition.org), and supported further by the G7 Leaders Declaration in May 2016 (www.whitehouse.gov/the-press-office/2016/05/27/g7-ise-shima-leaders-declaration) to 'recognize the importance of mitigating emissions of short-lived climate pollutants'. These opportunities include (i) venting and flaring of methane in coal-mines, while also improving worker safety, (ii) detecting and removing natural gas leaks, from wellpads upstream through the distribution chain downstream (e.g., McKain et al 2015), (iii) covering landfills, which reduces methane emissions while producing biogas for energy and transport usage, and (iv) developing farm bio-digesters, which has been extensively applied in Germany and is spreading to other European countries (e.g., Lebuhn et al 2014). Other strategies are being developed but need more research on potential unintended consequences. For example, modifying ruminants' diet (e.g., linseed fed) to limit methane emissions is currently being examined but needs evaluation against the quality of meat and milk (e.g., Marette and Millet 2014) and against emissions of other greenhouse gases such as N₂O. Modification of rice agriculture practices (e.g., semi-inundated paddies, dry cultivation) is well tested and promising, assuming yield and quality of the staple food for more than 3 billion people can be guaranteed (e.g., Sun et al 2016). Such mitigation policies in the agriculture and waste sectors are key to reducing methane emissions in most of the high emitting regions (figure 2).

Methane appears to play an increasing role in on-going anthropogenic climate change, particularly in light of the slowdown of CO₂ fossil fuel emissions over the past three years (figure 1, bottom right). Methane emissions from increasing agricultural activities seem to be a major, possibly dominant, cause of the atmospheric growth trends of the past decade (e.g., Herrero et al 2016). The rapid increase in methane concentrations offers a growing mitigation opportunity, acknowledging the need to balance food security and environmental protection (Wollenberg et al 2016). Keeping global warming below 2 °C is already a challenging target, with most of the attention placed primarily on CO₂ emissions. Such a target will become increasingly difficult if reductions in methane emissions are not also addressed strongly and rapidly.

The authors acknowledge the many scientists and funding agencies whose efforts and support contributed to the new Global Methane Budget released by the Global Carbon Project (globalcarbonproject.org). JGC thanks the support of the NESP-ESCC Hub.